BLIIPS

(Note: this page is always in need of updates)

I'm fascinated by books and libraries. Since recently I am a representative from my research group for the VU library, and I'm in the starting stage of an actual research project combining A.I. and libraries.

Project BLIIPS stands for Books and Libraries: Intelligence and Interaction through Puzzle- and Skinnerboxes. In this project, which is partially funded by Stichting Leenrecht and embedded in ACBA and DSA, I try to investigate a combination of two things: i) the introduction of artificial intelligence in the physical library to come up with future versions of the (digital) library, and ii) finding new business models for the physical library, aiming at innovation in this sector. Both aspects (A.I. and Algorithmology) fit in a more general question: "what happens to the (physical) library in our digital age?".

Oh... and yes... I built this picture myself (in all respects, that is). :)

This web-space will be your primary place for more information about the project, updates, results and all that. Some initial announcements can be found at leenrecht and VU-FEWEB.

At this moment I'm active with several sub-projects in BLIIPS:
1) Finishing and testing (together with an CS MSc student) the localization-navigation prototype app. In parallel, we are shifting some of the software building activities to a professional party to scale up.
2) Applying ethical frameworks for algorithms in the (public) library (see also my course on "ethics of algorithms" in the teaching section)
3) Empirically investigating the implications of the use of algorithms and big data in the (public) library. Together with a master student we are conducting interviews with various layers and people in the public library field (in the Netherlands) to find out what the status is of thinking about the (potential) consequences of algorithms and data on things like privacy, surveillance and freedom of reading in the public library of the future.
4) Surveying analog and digital approaches towards "measuring" activities of people in (public) libraries.
5) (Fundamental) Computational work on artificial intelligence algorithms for the "optimization" of various (public) library services.
6) Acquisition and information gathering the big field of public libraries (in the Netherlands so far) looking for "data projects", which can be about marketing, use of big data, profiling of customers, optimization of services and much more.

This was our (Dutch) poster for a couple of days empirical data gathering with our phones in the library

Running around with a prepared phone, just like our patrons had to do

Many things have been started, investigated or published already, see below.

I'd love to come into contact with interested people from universities and (public) libraries.

Forthcoming
I'll be an invited speaker at QQML 2017 on the interplay between technical algorithms and the ethical consequences of applying them in library contexts
van Otterlo, M. (2017) Technical fixes and ethical challenges for activity monitoring in physical public libraries through big data and artificial intelligence --- Invited talk at the 9th International Conference on Qualitative and Quantitative Methods in Libraries (QQML2017) in Limerick, Ireland (May 2017)

Forthcoming; download here.
van Otterlo, M. and Warnaar, M. (2017) Towards Optimizing the Public Library: Indoor Localization in Semi-Open Spaces and Beyond, proceedings of the Benelux Conference on Machine Learning (Benelearn-2017, to be presented June 2017)

A smartphone and a tablet with a beacon-powered localization app

Forthcoming
Warnaar, M (2017) Indoor Localisation on Smartphones Using WiFi and Bluetooth Beacon Signal Strength, MSc. thesis Vrije Universiteit Amsterdam, Computer Science Department

Just published
van Otterlo, M. (2016) Project BLIIPS: Making the Physical Public Library more Intelligent through Artificial Intelligence, in proceedings of the 8th International Conference on Qualitative and Quantitative Methods in Libraries (QQML2016)

The above paper is published in QQML journal

Picture shows five ways to interact with a book

Using advanced computer vision, categories on spines can be recognized with a smartphone camera.

Published. In this thesis we investigated several ways to interact with books, for example by scanning the front (and looking up the book on the internet), by scanning the chip in the book, by scanning the bar code, and by scanning the spine of the book (for example to look up the category of the book)
Jica, R.I. (2016), Digital Interactions with Physical Library Books, BSc thesis, Vrije Universiteit Amsterdam, Computer Science Department, July

Heatmaps of movement/occupation patterns of users on a floor can be computed using beacon localization

Published. In this thesis we focused on simple algorithms using only bluetooth beacons for localication, both in our university testing ground as in the public library.
Bulgaru, A. (2016) Indoor Localisation Using Bluetooth Low Energy Beacons, BSc thesis, Vrije Universiteit Amsterdam, Computer Science Department, December